Patients with clinical stage IB, IIA/B and IIIA non-small cell lung cancer (NSCLC) have poor 5-year survival rates that range from 15-40%. Preliminary results from an international cooperative study suggest that survival might be improved (approximately 5%) by treating patients with post-operative chemotherapy and/or radiation, although a high percentage of patients (23%) experienced serious Grade 4 toxicity. Adjuvant therapies that are more tumor-specific and less toxic are needed. We propose that dendritic cells (DC) loaded ex vivo with antigens from the patient's own irradiated autologous tumor, and matured with a combination of interferon-gamma (IFN-gamma) and inactivated/formaldehyde-fixed (BCG), can be used to safely induce a state of anti-tumor immunity and prolong tumor-free survival in the adjuvant setting. A Phase I clinical trial will be carried out to evaluate the safety and tolerability of administering this form of DC vaccine to postoperative patients with selected clinical stage IB, IIA/B or IIIA NSCLC. Tumor cells produce immunosuppressive factors that can suppress host immunity and enhance tumor survival. Resection of the primary tumor reduces these factors and temporarily restores immune responsiveness. Tumor resection also reduces tumor burden, reducing the risk that tumor variants will escape immune detection. As such, treating patients in the postoperative period provides an optimal window for employing anti-tumor vaccines. DC precursors will be harvested by leukapheresis and differentiated ex vivo with GM-CSF and IL-4. Autologous tumor will be harvested during the patient's primary tumor resection and purified using a novel immunodepletion protocol. Using the patient's own tumor, DC will be loaded with a full repertoire of tumor antigens capable of inducing both CD4 and CD8 T cell responses. After antigen loading, DC will be treated with BCG and IFN-gamma to enhance IL-12 production, co-stimulatory activity and capacity for T cell activation. The DC vaccine will then be administered as a monthly intradermal injection for three consecutive months following recovery from surgery. A total of 12-15 patients will be treated in two cohorts, one receiving 2 x 106 DC per immunization and the second receiving 6 x 106 DC per immunization. In addition to standard toxicity, tolerability and clinical outcomes, changes in serum cytokines (IL-10, VEGF, TGF-beta) and cell subsets (DC1, DC2, T suppressor) will be used to monitor changes in tumor-related immunosuppression; skin-test reactions to PPD, BCG-specific T cell proliferation and intracellular cytokine responses to DC loaded with BCG will be used to monitor responses to the vaccine; and T cell proliferation and intracellular cytokine responses to tumor-loaded DC will be used to monitor the development of tumor-specific immunity. Patients will be followed for a total of 1 year. Evidence that the vaccine is safe, documentation of a feasible manufacturing process, and evidence of vaccine-specific immune responses will provide the information required to support subsequent clinical trials focused on clinical outcomes and survival.